Case Study 3.2 — Exactly What You Asked For

A design-build distribution center, a criteria document with one missing sentence, and a $2,900,000 floor

Calder Foods, Ostrander Design Build, and every person named here are Tier-3 illustrative composites. The dispute pattern is real and common; the numbers are constructed to teach it.

Setup

Calder Foods is a regional grocery distributor operating out of two aging warehouses in the Rivermont metro. Its board approved a consolidated distribution center: 210,000 square feet, tilt-up concrete, 36-foot clear height, 42 dock positions, a small refrigerated section, and offices.

Calder had never built a building. It had a real estate director, Gwen Osei, who had done site acquisition and lease negotiation for eleven years and had never managed a construction project, and it had a distribution operations VP, Ted Brannigan, who knew exactly how his warehouse needed to run and had never been asked to write it down.

They chose design-build, for the two reasons owners usually choose it: they needed the building fast — the lease on the older warehouse expired in nineteen months — and they had nobody to manage a design team.

Their consultant produced a 34-page owner's criteria document and issued a request for proposals to five design-build teams. Four responded.

The criteria document was not bad. It specified clear height, bay spacing, dock count and door type, fire protection to the applicable NFPA standard for the commodity class, roof R-value, and a truck court dimension. It ran to 34 pages, which is respectable for a warehouse.

On the concrete slab, it said this:

"Concrete slab on grade shall be suitable for warehouse operations, minimum 8 inches thick, 4,000 psi."

That sentence is the case study.

What Happened

The bid that asked the question

Four proposals came in. Kestrel Construction Group, teamed with a design partner, bid it. Nadia Haddad, Kestrel's VP of Operations, ran the pursuit; Tomás Reyes estimated it.

Tomás flagged the slab language in the first review meeting.

Tomás: "It says suitable for warehouse operations. Suitable for what? There's no FF/FL number in here anywhere, and there's no material handling package. If they're running standard counterbalance forklifts in wide aisles, our number is fine. If they're going narrow-aisle with wire guidance at 36 feet, it isn't close."

Nadia: "Do we ask, or do we price it and clarify?"

Tomás: "We ask. And when they don't answer, we clarify."

They asked, in writing, during the question period. The answer that came back was a version of refer to the criteria document.

So Kestrel priced a slab consistent with ordinary random-traffic warehouse practice, and attached a clarification to its proposal:

"Criteria do not specify floor flatness or levelness tolerances, nor identify the material handling equipment package. We have priced floor tolerances consistent with random-traffic general warehouse use. If defined-traffic superflat tolerances are required for narrow-aisle wire-guided equipment, ADD $410,000 and 3 weeks."

Kestrel's proposal: $28,810,000. Ostrander Design Build's proposal: $28,400,000. No clarification on the slab.

Calder awarded to Ostrander. Kestrel lost by $410,000 — exactly the number it had disclosed.

The equipment nobody mentioned

Five months after the design-build contract was signed, Ted Brannigan's operations team completed its material handling study and selected the racking and equipment package: 38-foot-high very-narrow-aisle racking with wire-guided turret trucks.

Nobody told Ostrander. There was no reason anyone would have — the criteria document did not require the owner to furnish the equipment package, did not identify a racking vendor, and did not create any coordination obligation. The design-build contract was a fixed price for a building described in a document that never mentioned turret trucks.

Ostrander designed and built a code-compliant, well-executed 8-inch slab meeting ordinary random-traffic warehouse tolerances.

Commissioning

Fourteen months later, the racking went in and the first turret truck went down aisle 6.

At 38 feet of mast extension, small floor deviations translate into large deflections at the top of the mast. The trucks swayed enough that the operators would not run them at rated speed. Two rack uprights were struck in the first week. The equipment vendor measured the floor in the defined traffic aisles and reported that it did not meet the tolerances the equipment required.

The slab was not defective. It simply was not superflat, because nobody had ever asked for superflat.

The remedy

Plus 14 weeks of delayed full operation, during which Calder ran a partial operation out of the new building and held the old lease it had planned to release.

The Dispute

Calder's position. The criteria said "suitable for warehouse operations." A floor that cannot support the equipment used in a modern high-bay warehouse is not suitable for warehouse operations. Design-build exists so that the owner gets a result, and Calder did not get one. Ostrander is the expert; Calder is a grocery distributor.

Ostrander's position. The criteria specified thickness and strength and nothing else. Floor flatness in a warehouse is entirely dependent on the material handling equipment, which the owner selected five months after contract and never disclosed. Ostrander delivered a slab meeting ordinary industry practice for general warehouse use, at the tolerance any reasonable designer would assume absent a stated requirement. Had it known, it would have priced it — as one of its competitors expressly did.

Where it actually turned. Three things decided this, and none of them was the word "suitable."

1. The standard of care, not a warranty. Ostrander's design obligation was to exercise the care and skill ordinarily exercised by design professionals on similar projects. Calder's theory required something closer to a warranty of fitness for a purpose it had never stated. Its own expert conceded that a designer with no knowledge of the equipment package would ordinarily specify a random-traffic tolerance. That concession was the hinge of the case.

2. The insurance followed the legal theory. Ostrander's professional liability carrier engaged on the negligence theory and declined the warranty theory — professional liability policies generally respond to negligent professional services, not to contractual guarantees of outcome. That meant the insured portion of Calder's claim was much smaller than the claim itself, and Calder's practical recovery was capped by what Ostrander was willing and able to pay from its own balance sheet.

3. Kestrel's clarification became evidence. In discovery, Calder obtained the four proposals. Kestrel's clarification demonstrated that a reasonable bidder had identified the gap, raised it, and priced it — which cut both ways. It supported Calder's argument that a competent design-builder should have asked. It supported Ostrander's argument that the requirement was genuinely absent from the criteria, since a sophisticated bidder had to write a paragraph explaining that it was missing.

Settlement at mediation, twenty-two months after commissioning:

Calder's total exposure, including legal and expert fees on both the dispute and the extended lease, exceeded $2,400,000 — against the $410,000 it declined to pay on award day.

Analysis

The criteria document is the contract. In design-bid-build, the drawings are the contract and the owner warrants them. In design-build, the criteria are the contract and the owner warrants those. Calder did not escape the implied warranty of what it furnished; it just furnished 34 pages instead of 300, and it warranted those 34 pages. The risk did not disappear when the document got shorter. It concentrated.

"Suitable" is not a specification. Every soft word in a criteria document — adequate, suitable, industry standard, as required, appropriate — is an unpriced option granted to the reader. The design-builder will interpret it reasonably and cheaply, which is exactly what it is entitled to do. If a requirement can be stated as a number, state it as a number. If it cannot be stated yet because the owner has not decided, say so explicitly and create a mechanism: an allowance, a deferred package, an owner-furnished-information deadline with a stated consequence.

Owner-furnished information needs a date and a consequence. The single most useful clause missing from Calder's contract was one requiring Calder to furnish its material handling equipment package by a stated date, with a defined process — and a defined price adjustment — if the package arrived late or drove requirements beyond those in the criteria. Ostrander would have signed that clause happily. It just was not asked to.

The bidder who asks the right question loses. This is the uncomfortable structural lesson, and it is not confined to design-build. Kestrel did the professionally correct thing: it identified a genuine gap, raised it in the question period, and disclosed a price for the risk. It lost the job by exactly the amount of the disclosure. Any procurement that evaluates only the bottom-line number systematically rewards silence about risk.

The defense is on the owner's side of the table: read the clarifications. A single proposal carrying a clarification nobody else carried is not a bidder being difficult. It is free consulting, delivered by somebody who read your document more carefully than you did. Gwen Osei had Kestrel's paragraph in her hands nineteen months before the first turret truck went down aisle 6.

A note on what design-build did right. It is worth saying that Ostrander's building was good. The dock configuration worked, the fire protection was correct, the roof did not leak, and it was delivered on schedule for a fixed price. Design-build did what design-build does: it produced exactly what the criteria described, fast, for a price agreed at the start. That is not a failure of the method. It is the method operating precisely as designed on an input that was incomplete.

Discussion Questions

1. Rewrite the slab sentence so this dispute cannot happen. Your version must specify a measurable requirement, identify what the owner will furnish and by when, and state the consequence if it arrives late or changes the requirement. Three sentences maximum.

2. Was Ostrander's failure to ask about the equipment package a breach of the professional standard of care? Argue both sides. Does your answer change if Ostrander had built four narrow-aisle facilities in the previous five years?

3. Calder saved $410,000 on award day and spent more than $2,400,000 as a result. Design a procurement evaluation process — one that a first-time owner could actually run — that would have surfaced Kestrel's clarification as a decision factor rather than a footnote.

4. Under design-bid-build, whose problem would this floor have been? Under CM at risk? Work through each and be specific about the mechanism, not the outcome.

5. The insurance point in the analysis is subtle and expensive. Explain to a non-technical owner, in plain language, why negotiating a stronger warranty from a design-builder can leave the owner worse protected than a standard-of-care obligation.

Your Turn

You are Nadia Haddad. Kestrel just lost Calder Foods by $410,000, and the pursuit cost the firm about $60,000 in proposal and preliminary design time.

Write the internal debrief memo — half a page, to Kestrel's operations leadership — answering three questions:

1. Would you write that clarification again on the next design-build pursuit? Yes or no, and why.
2. Is there a way to have disclosed the risk and won the job? Describe it concretely — what would you have put in the proposal, and where?
3. What single screening question should Kestrel add to its design-build go/no-go checklist as a result of this pursuit? (Bid/no-bid decisions get their full treatment in Chapter 15.)

Then, separately, write the two-sentence version you would say to a young estimator who asks why Kestrel bothered.